An improved multivariate data visualization technique

With the heavily increase of information cognitive burden, it is urgent for us to find effective visualization tools to grasp the abundant multidimensional and multivariate information in the science, engineering and commerce fields. Star coordinates is a traditional multivariate data visualization technique, with some limitations, e.g. the loss of information, the poor result of visualization and the complexity of manual configuration of dimension axis. Advanced star coordinates (ASC), which addresses these drawbacks previous, is proposed in this work. ASC uses the diameter, but not the radius, as the dimension axis, and designs the dimension configuration strategy to optimize the order and angle of the dimension axes and project the multidimensional information object to low dimension visual space. Our experiment result tells that the dimension reduction algorithm is of great efficiency. Dimension configuration strategy reduces userspsila operation burden greatly and helps exploring the connotative characteristics of the multidimensional information aggregation quickly and exactly. The visualization result is easily understandable and expresses the dimension distribution information effectively, which is helpful for user to view the multidimensional data and to discover the implicit information in knowledge discovery process.